Test compiled with MinGW. Not runtime tested.
Adjusted the prototype to be in-one-line, as instructed. And have
ignored the other nits :-)
Your patch has been applied to the master branch.
commit bcf04b0b8e3c0f142ab0ec97627ea140c80c7962
Author: Lev Stipakov
Date: Mon Jan 17 11:49:17 2022 +0
I have no idea what this does, but I understand that it's related to
MSVC building and that we generally want/need to bump pkcs11-helper to
version 1.28 -> so, this is what it does :-)
Since this does not touch anything code or mingw-build-related I have
not tested it beyond "does 'git show' look
Acked-by: Gert Doering
As stated before, I have no idea what this does - but it doesn't touch
code or MinGW building (= not tested anything) and looks reasonable for
the purpose described.
It would be good to have 2.5 test installers with this, to verify that
PKCS#11 and EC do work now...
Your
Hi,
On Tue, Oct 19, 2021 at 08:31:27PM +0200, Arne Schwabe wrote:
> The signature messages required by external key managed also break
> the 1280 limit. To also avoid this surprise of different behaviour
> with PKCS11 enabled/disable, always use the larger size.
JFTR, this patch has been obsolete
In the past we had issues with patches accidentally breaking less common
OpenVPN configurations. For this reason it makes sense to some of those
configurations build via CI.
Adding:
* --enable-iproute2
* --enable-async-push
* --disable-management
* --enable-small
* --disable-lzo --disable-lz4
The
Acked-by: Gert Doering
Sounds useful. We should not overdo the "burn energy for testing build
variants" thing, but "not test problematic variants" is not good either.
Your patch has been applied to the master branch.
commit 8ee0997abd0642c4c4a6135573beba647d982f7c
Author: Antonio Quartulli
Dat
This unit-test did not consider the case when USE_COMP is not defined,
thus generating a compiler error.
Adapt the test to the case when no compression is available.
Cc: Arne Schwabe
Signed-off-by: Antonio Quartulli
---
v2:
* restore original MTU numbers - this can be addressed by another patc
Hi,
On Fri, Jan 21, 2022 at 03:09:41PM +0100, Antonio Quartulli wrote:
> This unit-test did not consider the case when USE_COMP is not defined,
> thus generating a compiler error.
>
> Adapt the test to the case when no compression is available.
>
> Cc: Arne Schwabe
> Signed-off-by: Antonio Quar
This unit-test did not consider the case when USE_COMP is not defined,
thus generating a compiler error.
Adapt the test to the case when no compression is available.
Cc: Arne Schwabe
Signed-off-by: Antonio Quartulli
---
v2:
* restore original MTU numbers - this can be addressed by another patc
Acked-by: Gert Doering
This looks better, and actually works :-)
Your patch has been applied to the master branch.
commit 33d9c7ace354693168c093550bd5a9f4dfed4078
Author: Antonio Quartulli
Date: Fri Jan 21 15:43:13 2022 +0100
unit-test: fix test_crypto when USE_COMP is not defined
Hi,
On Wed, Jan 19, 2022 at 07:21:26PM +0100, David Sommerseth wrote:
> index 5626e2b6..eb0b1254 100644
> --- a/src/openvpn/crypto.c
> +++ b/src/openvpn/crypto.c
> @@ -34,6 +34,7 @@
> #include "error.h"
> #include "integer.h"
> #include "platform.h"
> +#include "openssl_compat.h"
>
> #includ
From: Selva Nair
An easy way to trigger this error is to run an otherwise working setup
(at say verb = 4) with increased verbosity of verb >= 7 and using a GCM
cipher (e.g., AES-256-GCM). It will cause a fatal exit while printing the
cipher and hmac in key2_print().
Signed-off-by: Selva Nair
Hi
On Fri, Jan 21, 2022 at 12:10 PM Gert Doering wrote:
> Hi,
>
> On Wed, Jan 19, 2022 at 07:21:26PM +0100, David Sommerseth wrote:
> > index 5626e2b6..eb0b1254 100644
> > --- a/src/openvpn/crypto.c
> > +++ b/src/openvpn/crypto.c
> > @@ -34,6 +34,7 @@
> > #include "error.h"
> > #include "integ
From: David Sommerseth
With commit 544330fefedc87, the openssl_compat.h got included in
crypto.c. This caused issues when building against mbed TLS, which this
compat layer is not targeting.
This issue is resolved by only including this header when the OpenSSL
library is in use. The OPENSSL_FI
Signed-off-by: Antonio Quartulli
---
This patch combines some master commits in order to bring the GH script
up to the same state as the one in master (minus OpenSSL3 related
changes).
.github/workflows/build.yaml | 68
1 file changed, 61 insertions(+), 7 de
Hi,
On 13/12/2021 17:28, Frank Lichtenheld wrote:
The information provided by this is minimal
and it makes it more difficult to provide
deterministic builds. There are work-arounds
for that but I think it is easier to just remove
it completely.
Allows the build to pass with -Wdate-time -Werror.
Am 21.01.22 um 21:49 schrieb Antonio Quartulli:
Signed-off-by: Antonio Quartulli
---
This patch combines some master commits in order to bring the GH script
up to the same state as the one in master (minus OpenSSL3 related
changes).
Acked-By: Arne Schwabe
_
Am 21.01.22 um 19:57 schrieb selva.n...@gmail.com:
From: Selva Nair
An easy way to trigger this error is to run an otherwise working setup
(at say verb = 4) with increased verbosity of verb >= 7 and using a GCM
cipher (e.g., AES-256-GCM). It will cause a fatal exit while printing the
cipher and
18 matches
Mail list logo
